Rerun - The easiest way to build AI agents for all your tasks
by•
Most AI agents are black boxes. Rerun isn't. Build no-code agents that run 24/7, chasing invoices, qualifying leads, clearing your inbox, and watch every step in real time. They pause for approval before anything sensitive. Each workspace gets its own private server. Start with the included model, or connect your own API key or subscription
Replies
Rerun
@clement_janssens Curious about the failure mode more than the happy path. When we automated recurring ops work internally the problem was never building the thing, it was the agent quietly doing it wrong for three weeks before anyone noticed. What does Rerun surface when a run half-succeeds, and can I set a rule like "stop and ask me" on specific steps?
Rerun
@artem_fedorovich Totally agree with you Artem
Rerun gives tools to really understand the agent and stay in control
The agent has some handy features, like notifying the user and asking for approval before moving to the next step. Super useful for stuff like publishing an article or a LinkedIn post
We also built super detailed monitoring tools: steps, token usage (input/output), run times, etc., so you can always analyze and know what’s happening
Right now, most agentic solutions are black boxes,you just hope the output’s good. With Rerun, we’ve got agents you fully control
@clement_janssens Curious: for teams that need strict compliance or audit trails, what controls and visibility does Rerun provide around who can create/approve agent actions, how long activity logs are stored, and how easy is it to export those logs for audits? Also; what’s the simplest example you’d recommend for a small team to try in the first 24 hours?
Rerun
@etiennegarciaThank you, man! It’s definitely possible. You can also create your own personalized widget that aggregates all the tasks requiring your approval in one place.
Rerun
@etiennegarcia Exactly, especially during the first automation runs. Even if the skill is accurate, it's not perfect on the first try. Making sure you’ve got control before critical phases is key for safety and for iterating to get an agent that truly fits the use case
TypeBoost
Wow, this looks extremely powerful! Definitely have to check this out!
Rerun
@bennyqp Thanks Benny, we're super proud of the product, can't wait to get your feedback once you've tried it!
The hard part with always-on agents isn't watching them work, it's what happens after an approval. If a human approves step 4 and step 7 goes wrong, can you replay from the pause point with corrected context, or does the whole run start over?
Rerun
@aidan_codefox Totally agree.
Since the flows are unique, there's no way to "rollback."
However, the agent is smart enough to cancel changes if the user asks.
For example, with Rerun, any agent handling important processes has two key steps:
First, every critical or risky step requires user approval explaining exactly what it’s going to do. This ensures it hasn’t gone off track before taking action
Second, the agent notifies users once the step is done to confirm everything went smoothly.
If not, you can make it redo or cancel the changes
IMO the straight 'there's no rollback' answer is worth more than a polished one, most tools in this lane would have claimed replay works and hoped nobody tested it. QQ: when the agent cancels changes on request, is it walking back through the tool calls it made (voiding the invoice it created, deleting the record it wrote)? Congrats on the launch!
Rerun
@artstavenka1 Exactly, the agent knows what they’ve done which lets them undo changes (if possible)
That's why it's better for the agent to ask for approval BEFORE messing things up
one feedback, finding tools is slow, which i feel is not supposed to be, i feel you can fix that
Rerun
@daniel_oyetunde Hey Daniel, we’ll check it out right away, thanks for reporting it!
@clement_janssens great, because up till now is it still finding, i feel that will discourage some people Well done
watching every step is how you trust it early, but the real win is when you can stop watching. does the visibility taper as trust grows or stay always-on?
Rerun
@andrewzakonov You can keep track of and understand your agent throughout its entire lifecycle. If you ever have doubts, you can go check to ease your mind
The goal is to stay 100% in control of the agent, that’s the most important thing right now
Pushary
Lads, really cool stuff.
Love the design taste - congrats on the launch!
if you ever want to send push notifications on human in the loop steps to your end users, consider trying Pushary, we integrate with all the latest and greatest agent frameworks with a single line.
Pushary
https://pushary.com/human-in-the-loop
Rerun
@aadilghani Thanks Aadil
This is on the roadmap very soon. For now, it's email notifications only, but we're adding in-app and push notifications super soon
Appreciate the recommendation, I'll check it out!
"Watch every step in real time" is the part I care about most, because the black-box complaint is real. When a step fails halfway through a run, can you fix it and resume from that step, or does the whole agent re-run from the top? Resuming from the failed step is what makes long runs actually usable, and it is the hardest part to get right.
Rerun
@cmumulle This is honestly one of the things we're always working on
The agent is super independent, if a step fails, it'll find a way to make it work one way or another (like switching APIs or something)
But if it gets stuck, you’ll get notified and can fix it with the agent
Rerun sounds aimed at a pretty wide set of use cases, from Marketing automation to Engineering & Development and Productivity. How do you think about helping new users choose the right starting point for their first AI agent? Is the product more template-driven by category, or does it guide people from a task description into an agent setup?
Rerun
@ivory_xuxuxu One of the challenges with Rerun is that it's super flexible, so it can work for tons of use cases
We've put a lot of effort into a multi-step onboarding so you can run a 100% personalized agent without needing any tech skills
Feel free to try it out, would love to hear your thoughts